JP6861303B2 - How to operate the hearing aid system and the hearing aid system - Google Patents

Description

The present invention relates to a method for operating a hearing aid system including a first input converter, a second input converter and a signal processing device. The present invention further relates to a hearing aid system.

Hearing aid systems typically include one hearing aid, often with two hearing aids or are formed by two hearing aids. At that time, an old hearing aid that works to support the hearing impaired is usually called a hearing aid. In other senses, devices designed to assist a person who hears normally are also referred to by this term. Such hearing aids, also called "personal sound amplifiers" or "personal sound amplifiers" (shortly "PSAD"), are not provided to compensate for deafness, but are the hearing of ordinary people in special hearing situations. Used appropriately to support and improve. Used to assist hunters during hunting or observing animals, for example to better sense the sounds made by animals and other noises made by animals, and improved talk and / in more complex noises. Alternatively, it is used to support a sports reporter for the purpose of making it possible to understand words, and is also used to support a musician for the purpose of reducing the auditory load.

Regardless of the purpose of use described above, hearing aids generally include an input converter, a signal processor and an output converter as important components. Input transducers are typically formed by acoustic-electric-converters, eg, by microphones and / or by electromagnetic receivers, eg, induction coils. In most cases, electrical-acoustic transducers such as small speakers or electromechanical transducers such as bone conduction receivers are used as output transducers. The signal processing device is generally formed by an electronic circuit formed on a printed circuit board and usually includes an amplifier. The signal processing device processes the input signal generated by the input converter when the ambient sound is generated during the operation of the hearing aid, and functions to generate an output signal based on the input signal. This output signal is converted by an output converter, which makes it audible.

At that time, in order to process the input signal, it is advantageous to apply various algorithms according to the hearing aid situation at that time. This algorithm is suitable for a variety of expected hearing aid situations. In doing so, a characteristic of the expected individual hearing aid situation is, for example, the often repeated pattern of overlapping of the effective signal sound with interfering noise or general noise. In this case, the patterns are specifically categorized based on the type of noise generated, the signal-to-noise ratio, the frequency response of the active signal sound and / or the temporal variation and average of the above values.

At that time, the premise for the automatic conversion between various algorithms is the recognition of the hearing aid situation that exists each time, or at least the recognition of the change in the hearing aid situation that exists.

E. Hadad, S.A. Doclo and S. Gannot, "Binaural LCMV beamformer and it's performance analysis", IEEE Tran. On Audio, Sp, and Lang. Poc. , Aug. 2015 D. Marguardt and S.M. Doclo, "Performance Communication of Bilateral and Bilateral MVDR-based Noise Reduction Algorithms in the Presence of DOA Estimation Operations", in ITG Symposium, 2016, pp. 1-5 A. H. Kamkar-Parsi and M. Bouchard, "Improved noise power spectrum density restoration for binaural hearing aids operating in a diffuse noise field end vironment", IE. Audio, Speech, Lang. Process. , Vol. 17, no. 4, pp. 521-533, May 2009 R. Martin, "Noise power spectral estimation based on optimal smoothing and minimum statics", IEEE Trans. Speech Audio Process. , Vol. 9, no. 5, pp. 504-512, Jul. 2001

Starting from this, the underlying task of the present invention is to provide an advantageous method for operating the hearing aid system and an advantageously formed hearing aid system.

This problem is solved according to the present invention by the method having the feature of claim 1 and the hearing aid system having the feature of claim 12. Favorable developments are included in the dependent claims. The effects and advantageous embodiments listed in terms of method are similarly applicable to hearing aid systems and vice versa.

The method serves to activate hearing aid systems, especially the types of hearing aid systems mentioned at the beginning. This hearing aid system includes a first input converter, a second input converter, and a signal processing device. In doing so, in the process of implementing the method, the surrounding or surrounding area of the hearing aid simtem is monitored for the activity of the lateral effective signal source, so the method determines the activity of the lateral effective signal source around the hearing aid system. To.

This is because the first input signal is generated by the acoustic signal from the surroundings generated by the first input converter, the second input signal is generated by the acoustic signal generated by the second input converter, and the first input signal and the first input signal are generated. Based on the two input signals, a directional notch filter unit is used to generate a filtered input signal, and based on the further filtered input signal and based on the first and / or second input signal. The degree of attenuation caused by the directional notch filter unit is determined, and this degree of attenuation is contrasted with the reference, from which the presence or absence of lateral active signal source activity around the hearing aid system. Is done by guessing.

At that time, the method according to the present invention is particularly based on the basic idea of comparing two damping actions with each other. At that time, one of the two attenuation effects indicates the attenuation effect of the signal to be inspected, that is, particularly the first input signal and / or the second input signal by a kind of fade-out of the spatial angle range. In this spatial angle range, an active lateral effective signal source is inferred. This first damping action is compared to the second damping action that occurs when only part of the diffused background noise is faded out by fading out the spatial angle range. If the first damping effect is significantly greater than the second damping effect, then the activity of the lateral active signal source can be considered to be present, otherwise the activity of the lateral active signal source is not present. Can be regarded as.

In this case, the person with whom the person is speaking, that is, the person who speaks at least temporarily in the direction of the wearer of the hearing aid system, is generally regarded as an effective signal source. Such an effective signal source is when the wearer of the hearing aid system is looking straight ahead and not looking towards the active signal, that is, the effective signal source is away from or beside the wearer of the hearing aid system. When deviated, it is a lateral effective signal source. The effective signal source in the line-of-sight direction of the wearer of the hearing aid system is hereinafter referred to as the central effective signal source.

This distinction between the central and lateral active signal sources and the confirmation of when the lateral active signal source is active, that is, when the laterally offset speaker speaks, are especially important for hearing aid systems. This is advantageous when the wearer is talking to more than one person, thereby alternating between different active signal sources. By confirming the activity of such a lateral effective signal source, for example, a first and / or a second for the generation of an output signal or an exit signal, depending on whether the lateral effective signal source is operating or not. The processing of the input signal of is different.

The degree of attenuation described above is contrasted with the criteria described above to allow confirmation of the presence or absence of activity of the lateral active signal source around the hearing aid system. That is, a comparison is made between the determined degree of attenuation and the reference. This is done, for example, by finding the ratio of the degree of attenuation to the reference. In this case, in general, only the ratio or the value of the ratio is determined to be large or small.

Corresponding to other embodiments, a difference is determined, which determines whether the difference or value is greater than or less than zero, or greater than or less than a set threshold. For example, when comparing the degree of attenuation with the reference, if it is confirmed that the degree of attenuation is significantly smaller than the reference, the presence of activity of the lateral effective signal source is determined. On the other hand, if the degree of attenuation is greater than the reference, the absence of activity of the lateral active signal source is determined.

At that time, preferably, the attenuation coefficient or the logarithmic amount of attenuation is determined as the degree of attenuation. In this case, the damping coefficient or logarithmic damping amount is generally time dependent. It is even more advantageous if the reference indicates damping coefficient or logarithmic damping. In this case, the damping coefficient or the logarithmic damping amount is also generally time-dependent. Thereby, it is advantageous to compare the two damping coefficients or the two logarithmic damping amounts with each other.

In this case, first of all, the filtered input signal is generated to determine the degree of attenuation. For that purpose, a directional notch filter unit is used. The filtered input signal preferably matches one or more input signals of a hearing aid system with varying directivity, at least in good approximation. In this case, the directivity is formed so that a predetermined spatial range or spatial angular range in which the potential activity of the effective signal source is determined is substantially faded out. As a result, the portion of the acoustic signal from this spatial angle range is largely neglected.

To that end, it is determined in which direction the potential activity of the active signal source should be directed with respect to the hearing aid system, and around the corresponding direction or associated angular position to generate the filtered input signal. The set or short angular range of, eg, 10 ° angular range, is faded out. However, is the range around the central angular position, that is, the range around the line-of-sight direction of the hearing aid system wearer when looking straight ahead, removed in determining the potential activity of the active signal source? Or not considered. The determination of the potential activity of the active signal source is made, for example, when the set threshold of the signal level in the spatial angle range is exceeded. At that time, it is advantageous because the reference angle position, that is, the angle position 0 ° is not necessarily determined at the above-mentioned central angle position, that is, in the line-of-sight direction of the wearer when looking straight ahead.

The directional notch filter unit is considered to be known at least with respect to the basic principle (adaptive special notch beamforming). At that time, it is extremely important that there are two types. In the case of the first type, the so-called "binaural minimal dispersion distortion-free response beamforming (MVDR)" method is used. In the case of the second type, the so-called "binaural linear conditional minimum dispersion beamforming (LCNV)" method is used. The first type is described in detail in, for example, Non-Patent Document 1. The preferred second type is described in detail, for example, in Non-Patent Document 2.

Further, in an advantageous variant of the method, the reference is not easily set, for example in the form of a reference value, but is determined by determining the spectral output density of the disturbing noise based on the first and / or second input signals. Or determined by finding a value derived from the spectral output density for this disturbing noise. In the present application, it is advantageous that background noise generated from a person is regarded as interfering noise. This person is not talking to the wearer of the hearing aid system, for example talking to another person. Thus, disturbing noise includes, for example, so-called background chatter in cafeterias or open spaces. Such background or jamming noise is generally generated as diffused jamming noise, i.e., jamming noise that cannot be uniquely assigned to a source in place and is not directly directed towards the wearer of the hearing aid system. To do.

An advantageous method for determining such a spectral output density of disturbing noise is described in detail in Non-Patent Document 3. Alternative methods are described, for example, in Non-Patent Document 4.

It is advantageous that the value derived from the spectral output density of the interfering noise is the actual output of the interfering noise, the actual output value of the interfering noise, or the actual output average value of the interfering noise. This is an alternative to the disturbing noise output that can be derived from the first and / or second input signals over a generally set time and usually a set frequency band.

The actual output value of the disturbing noise is determined, for example, during the set first time interval, for example, during the first time interval of about 10 ms and during the set frequency band. At that time, it is purposeful that the set frequency band is oriented toward human speech. In this case, it is not always necessary to cover the entire frequency spectrum of human speech from about 80 Hz to about 12 kHz. In some cases, instead, a frequency band is set that includes frequencies from about 100 Hz to about 500 Hz. It is advantageous to consider the frequency band from about 125 Hz to about 4 kHz.

It is even more advantageous to determine the actual output value of the disturbing noise at a set second time interval, eg, a second time interval of about 100 ms, and generally the actual disturbing noise determined. Since each output value remains valid for the time of the set second time interval, it is possible from this to derive the time course of the actual output of the disturbing noise over the set frequency spectrum, preferably derived. Is done.

In another advantageous way, the frequency components considered are weighted, eg, a weighted mean is determined. This is done in particular on the basis of frequency components over the frequency band from about 125 Hz to about 4 kHz.

In a favorable development, the directional notch filter unit is further used to determine the parameter value for at least one correction parameter, or particularly by forming the directional notch filter unit, the appropriate parameter value for at least one correction parameter. Is set. The at least one correction parameter or the plurality of correction parameters is, in particular, the adaptive filter coefficient of the directional notch filter unit. At that time, the number of correction parameters usually matches the number of channels used or input signals.

Corrected spectral output density or correction of disturbing noise based on or derived from the spectral output density of the disturbing noise and by the parameter values of at least one correction parameter or by multiple parameter values of multiple correction parameters. It is even more advantageous if the derived value is determined. That is, for example, starting from the time course of the actual output of the disturbing noise over the set frequency spectrum, the time course of the corrected actual output of the disturbing noise over the set frequency spectrum is determined. It is advantageous.

For example, consider the determined spectral output density S of the disturbing noise and the correction parameters P1 and P2. This correction parameter indicates the adaptive filter coefficient of the directional notch filter unit. In particular, the parameter values P1 and P2 change depending on the spatial position of the notch of the directional notch filter unit. In this case, ^{the corrected spectral output density of the disturbing noise S *} can be obtained from, for example, the following relational expression.

^{S * = (| P1 | ∧} 2+ | P2 | ∧ 2) S

For example, first of all, the temporal passage of the actual output of the disturbing noise over the set frequency spectrum is derived from the first input signal and / or the second input signal. In order to determine the corrected derivation value, that is, the corrected actual output of the disturbing noise, the disturbing noise output is always evenly distributed in all spatial directions, which is expected in the case of diffuse background noise. Assumed. In this case, one parameter value for at least one correction parameter or multiple parameter values for the plurality of correction parameters is a spatial range that is faded out, for example, by a directional notch filter unit to determine the degree of attenuation. Represents width or size. Finally, this information derives a modified actual output of the jamming theory from the actual output of the jamming noise. This output is part of the actual output of disturbing noise, which is faded out by the spatial range fade-out by the directional notch filter unit.

Further, in an advantageous modification method, the spectral output density of the interfering noise or the modified spectral output density of the interfering noise is compared to the overall spectral output density to determine the reference. In this case, the overall spectrum output density is determined based on the first input signal and / or the second input signal. Instead, to determine the reference, the value derived from the spectral output density of the disturbing noise, the derived modified value of the disturbing noise, or the value derived from the corrected spectral output density of the disturbing noise is the entire spectrum. Contrast with the value derived from the output density. The overall spectrum output density is advantageous because the overall output is easily considered from the first input signal and / or the second output signal.

In the modified embodiment, the reference indicates, for example, the attenuation of the actual total output when the actual total output is reduced by only the corrected actual output of the interference noise. At that time, the actual total output is determined in the same manner as the actual output of the disturbing noise. That is, the same time interval and the same frequency band are set, but the overall output is considered from approximately the first input signal and / or the second output signal. That is, it is based on the output density of the entire spectrum. The reference or actual reference indicates the actual attenuation coefficient or the actual logarithmic amount of attenuation.

Further, in order to determine the degree of attenuation, it is desirable to determine the spectral output density of the filtered input signal and compare it with the overall spectral output density, especially the overall spectral output density described above. The overall spectrum output density is determined based on the first input signal and / or the second input signal.

Furthermore, when determining the degree of attenuation, it is advantageous to operate at the derived values, especially at the actual output. Therefore, in order to determine the degree of attenuation, it is advantageous to compare the actual output of the filtered input signal with, in particular, the actual overall output that matches the actual overall output described above. At that time, in order to make comparisons possible, for example, the same time interval and the same frequency band as in the case of the actual output of interfering noise are set. In this case, the degree of attenuation or the actual degree of attenuation indicates the actual attenuation coefficient or logarithmic actual amount of attenuation.

Given that the degree of attenuation and the reference indicate the actual damping coefficient or the actual logarithmic amount of attenuation, respectively, the degree of attenuation and the reference can be easily compared and contrasted with each other, for example by finding the difference. To that end, for example, the degree of attenuation or the actual degree of attenuation and the reference or the actual reference are supplied to the comparison unit. At that time, it is advantageous for the comparison unit to output a binary determination signal having two values. In this case, one value represents the presence of activity of the lateral active signal source and the other value represents the absence of activity of the lateral active signal source.

In a favorable evolution, an offset value is further set for the comparison unit. The determination threshold is shifted by this offset value. This is advantageous in determining the difference between the degree of attenuation and the reference to change the output signal of the comparison unit. That is, for example, it is advantageous to determine how much the degree of attenuation is greater than or less than the reference. Thereby, the presence of activity of the lateral active signal source can be determined. In doing so, changes in offset values generally shift the compromise between sensitivity and error proneness towards sensitivity or error proneness.

As described above, the perimeter of the hearing aid system is monitored based on the activity of the lateral active signal source by the above method or by the above portion of the method according to the invention. In doing so, surveillance makes it possible to confirm the presence of activity of the lateral active signal source, which is used in a favorable development form for the control of the hearing aid system and especially for the activation or start of auxiliary functions. To. In this case, it is advantageous that the auxiliary function is activated and as a result implemented when the activity of the lateral active signal source is determined around the hearing aid system. It is advantageous for activity confirmation to function as a kind of trigger. Whenever the activity of a lateral active signal source is determined around the hearing aid system, the trigger initiates an auxiliary function.

At that time, the auxiliary function confirms that an appropriate hearing aid program is selected or the presence of lateral active signal source activity depends on the actual hearing aid situation, corresponding to the advantageous modification embodiment. It is easily switched between the two hearing aid programs, depending on whether it is confirmed to be absent or not. That is, for example, when the absence of lateral active signal source activity is determined, the listening system is activated by the first hearing aid program, and the presence of lateral active signal source activity is determined. The listening system is activated by the second hearing aid program.

In addition, method modification is advantageous. In this method modification, a signal processing device is used to generate an output signal depending on at least one parameter value of at least one parameter for signal processing, and an auxiliary function according to this output signal causes an actual hearing aid situation. At least one parameter value is fitted to. At that time, for example, so-called beamforming is performed based on at least one parameter value, and the directional characteristics of the hearing aid system are generally adapted by matching the at least one parameter value.

In addition, method modification is advantageous. In the case of this method modification, an auxiliary function is used to determine the relative location or position of the laterally active signal source with respect to the hearing aid system. This relative location or position indicates the direction in which the lateral effective signal source should be placed, especially in relation to the line-of-sight direction of the hearing aid system wearer when looking straight ahead. In a favorable evolutionary form, the relative location or relative position is not only determined once, but instead the relative location or relative position of the laterally active signal source is continuous, if possible. Pursued in time.

The method according to the present invention described above is useful, for example, for operating a hearing aid system, and is therefore for a hearing aid system, as described above. The hearing aid system according to the present invention adapts to at least one operating mode to carry out the above method, and includes a first input converter, a second input converter, and a signal processing device. During the operation of the hearing aid system, the first input converter produces the first input signal and the second input converter produces the second input signal. The first input signal and / or the second input signal is not used only for carrying out the method according to the present invention described herein, depending on the modified embodiment of the hearing aid system. Instead, both input signals, i.e. the first input signal and the second input signal, can generally supply one input signal or both input signals in parallel to a plurality of signal processing processes as needed. Be prepared.

The above-mentioned principle for this signal processing can be realized depending on whether an analog signal exists and analog signal processing is performed or whether a digital signal exists and digital signal processing is performed. Is. That is, the above-mentioned first input signal and the above-mentioned second input signal are analog signals or digital signals according to a modified embodiment of the method according to the present invention or according to the realization of the method according to the present invention. is there. However, in the case of signal processing, a digital signal is advantageous, and a digital signal processing is advantageous. This digital signal processing is performed, for example, by a microprocessor, which is a particular part of a signal processing device. The above partial steps of the method are usually performed or implemented by logical or virtual elements.

Regardless of whether analog signal processing or digital signal processing is performed, the time between the change in the activity of the laterally active signal source, that is, the start or end of the activity, and the detection of the change by the hearing aid system. It is advantageous that the hearing aid system is formed so that the delay is less than about 100 ms.

It is purposeful that the hearing aid system further comprises a first hearing aid and a second hearing aid. At that time, it is advantageous that the first input converter is a part of the first hearing aid and the second input converter is a part of the second hearing aid. Instead, the first input converter and the second input converter may be part of the first hearing aid.

In a slightly modified embodiment, the hearing aid system further comprises one or more other input converters in addition to the first input converter and the second input converter, by the other converter, the first input. In addition to the signal and the second input signal, other input signals are generated. It is advantageous if other input signals are additionally used to determine the reference and / or degree of attenuation. In doing so, for example, the proximity detector of the hearing aid system is used as another input transducer and to generate other input signals.

Next, an embodiment of the present invention will be described in detail with reference to a schematic diagram.

It is a block diagram of a hearing aid system. It is a plan view which shows the hearing aid situation with three talking companions, and in this case, one of the talking companions is wearing a hearing aid system. It is a graph which shows the temporal change of the acoustic signal from the hearing aid situation. It is a graph which shows the degree of attenuation, the reference and the time change of an output signal determined by a hearing aid system.

In all figures, the parts that match each other have the same reference numerals.

The hearing aid system 2, which is shown as a block diagram in FIG. 1 and will be illustrated below, is preferably formed as a binaural hearing aid system 2, and is intended to include a first hearing aid 4 and a second hearing aid 6. is there. In the embodiment, during use, the first hearing aid 4 is worn by the wearer 8 in the left ear, while the second hearing aid 6 is worn in the right ear.

In this case, the first hearing aid 4 includes a first input converter 10. During operation, the first input signal ES1 is generated by the acoustic signal AS generated by the first input converter 10. At that time, first, an analog signal is generated, and this analog signal is converted into a digital signal by the first A / D converter 12, and is provided to the signal processing device 14 as the first input signal ES1 in this form. In this case, the signal processing device 14 usually includes a microprocessor or computer chip or is formed by an electronic module.

The second hearing aid 6 itself comprises a second input converter 16, and like the first hearing aid 4, the second input signal ES2 is generated by the acoustic signal AS generated by the second input converter 16 during the operation of the second hearing aid 6. To generate. In this case, first of all, an analog signal is generated, and this analog signal is converted into a digital signal by the second A / D converter 18 and served as the second input signal ES2. The second hearing aid 6 further includes a second transmission / reception unit 20, which sends a second input signal ES2 to the first hearing aid 4 where it is received by the first transmission / reception unit 22. Since the second input signal ES2 is supplied from the first transmission / reception unit to the signal processing device 14 of the first hearing aid 4, the signal processing device 14 is provided with the first input signal ES1 and the second input signal ES2.

In an embodiment, the signal processing apparatus 14 implements the method according to the invention in at least one operating mode. This mode of operation determines the activity of the lateral active signal source 24 around the hearing aid system 2. At that time, the first hearing aid 4 worn on the left ear mainly monitors the left half space when viewed from the wearer 8, and the second hearing aid 6 worn on the right ear mainly monitors the right half space. Be monitored. That is, even if not shown, the second hearing aid 6 also includes a signal processing device. Further, since the first hearing aid 4 simultaneously sends the input signal ES1 to the second hearing aid 6, the signal processing device of the second hearing aid 6 is also provided with both input signals ES1 and ES2. Then, in both hearing aids 4 and 6, the method according to the present invention described below is carried out, respectively. Both hearing aids 4 and 6 simultaneously carry out the method according to the present invention in parallel.

At that time, we start from the hearing aid situation as shown in FIG. Here, the wearer 8 of the hearing aid system 2 is shown substantially in the center of the lower range of the figure. The line-of-sight direction of the wearer when looking straight ahead is 26 in the central direction. In front of the wearer 8 in the central direction 26, there is a first talking partner as a central effective signal source 28. The first talking partner is shown in the upper center in FIG. 2, which shows the hearing aid situation in a plan view. Somewhat to the left of that is the second conversation partner. This second talk partner is located laterally 30 when viewed from the wearer 8. The lateral direction 30 and the central direction 26 form an angle of about 70 ° in the present embodiment. As a result, the second talking partner is in a lateral or lateral position when viewed from the wearer 8 at least in the line-of-sight direction of the central direction 26 when looking straight ahead. The method described below functions to recognize when the second talking party, which is the laterally effective signal source 24, is just speaking, that is, when the laterally effective signal source 24 is active.

Therefore, the first input signal ES1 and the second input signal ES2 are processed by the signal processing device 14, and in particular, the first input signal ES1 and the second input signal ES2 are provided in parallel to the plurality of signal processing elements 32. .. That is, a plurality of the elements 32 can use both input signals ES1 and ES2 independently of each other, and both signals can be used as a base for the signal processing process.

In doing so, the various signal processing elements 32 are generally not formed by various 4-terminal networks or other electronic modules, but are realized by virtual units, eg, by various programs or processes that can be implemented in parallel. .. In the embodiment, at that time, the attenuation degree element 34, the reference element 36, the comparison unit 38, the directional notch filter unit 40, the first auxiliary element 42, and the second auxiliary element 44 are formed as the signal processing element 32.

In the directional notch filter unit 40, a filtered input signal GS is generated based on the first input signal ES1 and the second input signal ES2. Therefore, the directivity is simulated. Due to this directional characteristic, a predetermined spatial angle range around the lateral direction 46, for example, a spatial angle range of 10 ° around the lateral direction 46, which is indicated by two broken lines arranged on both sides of the lateral direction 46 in FIG. 2, is faded out. Therefore, the components of the acoustic signal AS generated from this spatial angle range are erased or faded out. The filtered input signal GS no longer displays the corresponding component.

In this case, however, the source direction 46 is not set constant, but is determined by a unique process that changes over time and progresses in parallel, and in particular the source direction 46 is a potential lateral effective source. Is determined to indicate the direction of. Therefore, strictly speaking, the source direction 46 is the actual source direction 46 or the time-varying source direction 46. For this purpose, first of all, an auxiliary signal is generated starting from the first input signal ES1 and the second input signal ES2. In addition, the directivity is simulated. Since this directivity fades out a set spatial angle range around the central direction 26, for example, a spatial angle range of 10 ° around the central direction 26, a part of the generated acoustic signal AS emitted from this spatial angle range. Is canceled or faded out. The corresponding part of the auxiliary signal is no longer displayed. In the remaining spatial range, the direction in which most of the generated acoustic signal AS reaches the hearing aid system 2 is sought. This direction is determined as the source direction 46. At that time, in the case of good proximity, the source direction 46 coincides with the lateral direction 30 whenever the lateral active signal source 24 is active.

Once the actual source direction 46 is determined, the parameter value P, which depends on the actual source direction 46, is calculated or derived for the parameter. With this parameter value, the above directivity can be simulated.

The parameter value P causes the first input signal ES1 to undergo a filter wave process. Thereby, the filtered input signal GS is obtained. In parallel with this, in a similar manner, the second input signal ES2 undergoes a filter wave process in the second hearing aid 6 by the parameter value P. That is, both input signals ES1 and ES2 are generally used to determine the source direction 46 and the parameter value P, but from one of the two input signals ES1 and ES2, the first input in the first hearing aid 4 It is advantageous if the filtered input signal GS is derived from the signal ES1 or the second input signal ES2.

Then, in the attenuation degree element 34, the time-dependent degree of attenuation M is determined based on the first input signal ES1 and the filtered input signal GS. In this case, the time-dependent degree of attenuation M indicates a logarithmic amount of attenuation. The First Therefore, on the basis of the first input signal ES1, actual overall output _{P G (ES1, Δt 1,} Δt 2, Δf) is determined. This overall output indicates the output of the acoustic signal AS that can be derived from the first input signal ES1 for the set first time interval Δt _{1 and the set frequency band Δf.}

At that time, it is a purpose that the set frequency band Δf corresponds to the human language. In this case, it is not always necessary to cover the entire frequency spectrum of human language from about 80 Hz to about 12 kHz. Instead, it is advantageous to set a frequency band that includes frequencies from about 125 Hz to about 4 kHz. At that time, it is more advantageous if the individual frequency portions are weighted. That is, for example, a weighted average value is obtained. For the first time interval Δt ₁ , for example, a time interval of 10 ms is set. Thereby, the _{output value can be determined for each time interval of the value Δt 1} , and the corresponding output value is determined at the set second time interval Δt ₂ , for example, the interval of the second time interval Δt ₂ of 100 ms. it is, and generally, since each output value determined to start from is constant during the time the time interval value Delta] t _2, now, the entire output _{P G (ES1, Δt 1,} Δt 2, Δf) It is advantageous because the time course of about can be derived over a set frequency spectrum.

Similarly, the attenuated first output _PD1 (GS, Δt ₁ , Δt ₂ , Δf) is also determined based on the filtered input signal GS. The degree M = M attenuation that depends on the time (t) following contrast M is _{(t) = 10dBlg [P D1} (GS, Δt 1, Δt 2, Δf) / P G (ES1, Δt 1, Δt 2, Δf )]]
It arises from.

At this _{time, P D1 (GS, Δt 1} , Δt 2, Δf) and _{P G (ES1, Δt 1,} Δt 2, Δf) a first value for the after-start t = 0 s of the method according to the present invention, Determined after a certain period of time.

In parallel with this degree of attenuation M, time-dependent reference R = R (by the signal treatment device 14 and based on both input signals ES1, ES2, i.e., first input signal ES1 and second input signal ES2. t) is determined. Therefore, first of all, in order to identify the diffused interference noise, both input signals ES1 and ES2 are evaluated together in the first auxiliary element 42, and the first interference signal S is determined. This first interfering signal has a component of the first input signal ES1 indicating diffused interfering noise. The first jamming signal S thus determined is then provided to the second auxiliary element 44. Similarly, in the second hearing aid 6, the second interfering signal is determined in parallel. This second interfering signal has a component of the second input signal ES2 indicating diffused interfering noise.

In the second auxiliary element 44, in order to obtain the filtered input signal GS, the first interfering signal S undergoes the same filter wave process as the first input signal ES1 by the parameter value P. Thereby, the modified first interference signal MS is obtained. This modified first jamming signal MS is provided to the reference element 36.

In the reference element 36, a time-dependent reference R is determined. The time-dependent criterion also indicates the amount of logarithmic attenuation. _{Therefore, the attenuated second output PD2} (MS, Δt ₁ , Δt ₂ , Δf) is determined based on the modified first jamming signal MS. In this case, further, the frequency band Δf set as before and the time intervals Δt ₁ and Δt ₂ set as before are used. The time-dependent criterion R = R (t) is
_{R (t) = 10dBlg [P} D2 (MS, Δt 1, Δt 2, Δf) / P G (ES1, Δt 1, Δt 2, Δf)]
It arises from.

Finally, a time-dependent degree of attenuation M and a time-dependent reference R are supplied to the comparison unit 38 where they are compared to each other. If the degree of time-dependent attenuation M is much smaller than the time-dependent criterion, the presence of lateral active signal source activity is determined, otherwise the lateral active signal source activity is non-active. Existence is determined. At that time, the comparison unit 38 generates a binary determination signal E having, for example, values 0 and 1. In this case, a value of 1 indicates the presence of active signal source activity and a value of 0 indicates non-existence.

The time course of the degree of attenuation M, the time-dependent reference R and the associated decision signal E is shown in FIG. At that time, for the set time intervals Δt ₁ and Δt ₂ , for example, a time interval smaller than the above 10 ms or 100 ms is used. Further, the offset value O is taken into account. This offset value changes the value of the determination signal E to the value 1 only when the difference between the degree of attenuation M and the reference R is equal to or greater than the set value.

For comparison, FIG. 3 shows the relevant temporal course of signal levels. This signal level indicates the acoustic signal AS or the strength of the acoustic signal AS. Furthermore, it indicates at what time the lateral active signal source 24 is active, that is, t = 3s to t = 6s and t = 10s to t = 13s, and the central active signal source. It shows at what time 28 is active, that is, from t = 6s to t = 10s and from t = 10s to t = 13s. The diffused disturbing noise is persistently present in the illustrated time interval.

It is advantageous that the decision signal E activates or deactivates the auxiliary function or switches it, for example, between two programs.

2 Hearing system 4 1st hearing aid 6 2nd hearing aid 8 Wearer 10 1st input converter 12 1st A / D converter 14 Signal processor 16 2nd input converter 18 2nd A / D converter 20 2nd transmission / reception unit 22 1st transmission / reception unit 24 Lateral effective signal source 26 Central direction 28 Central effective signal source 30 Lateral direction 32 Signal processing element 34 Damping degree element 36 Reference element 38 Comparison unit 40 Directional notch filter unit 42 1st auxiliary element 44 2nd auxiliary element 46 Source direction AS acoustic signal ES1 1st input signal ES2 2nd input signal GS Waved input signal P Parameter value M Degree of attenuation R Criteria S 1st interference signal MS Corrected 1st interference signal E decision signal O offset

Claims (14)

A method for operating a hearing aid system (2) including a first input converter (10), a second input converter (16), and a signal processing device (14).
The first input signal (ES1) is generated by the acoustic signal (AS) generated by the first input converter (10), and the second input is generated by the acoustic signal (AS) generated by the second input converter (16). A signal (ES2) is generated and
Based on the first input signal (ES1) and the second input signal (ES2), a filtered input signal (GS) is generated by using the directional notch filter unit (40).
Attenuation caused by the directional notch filter unit (40) based on the filtered input signal (GS) and based on the first input signal (ES1) and / or the second input signal (ES2). The degree (M) of is determined, and this degree of attenuation (M) is contrasted with the reference (R), from which the presence or absence of activity of the lateral active signal source (24) in the surroundings is inferred. By being done
The method described above, wherein the activity of the lateral active signal source (24) around the hearing aid system (2) is determined. The reference (R) is determined by determining the output density of the spectrum relating to the disturbing noise based on the first input signal (ES1) and / or the second input signal (ES2). The method according to claim 1. A correction parameter is derived by the directional notch filter unit (40), and a corrected spectral output density for interference noise is determined based on the spectral output density for interference noise and based on at least one correction parameter, or said. The directional notch filter unit (40) determines the parameter value (P) for at least one correction parameter and is related to interference noise based on the spectral output density for interference noise and by the parameter value (P) for at least one correction parameter. The method of claim 2, wherein the modified spectral output density is determined. To determine the reference (R), the spectral output density for interfering noise or the modified spectral output density for interfering noise is compared to the overall spectral output density, and this overall spectral output density is the first input signal ( The method according to claim 2 or 3, wherein the method is determined based on the ES1) and / or the second input signal (ES2). In order to determine the degree of attenuation (M), the spectral output density for the filtered input signal (GS) is determined and compared to the overall spectral output density, and this overall spectral output density is the first. The method according to any one of claims 1 to 4, wherein the method is determined based on the input signal and / or the second input signal. By supplying the degree of attenuation (M) on the one hand and the reference (R) on the other side to the comparison unit (38), the degree of attenuation (M) is compared with the reference (R). The method according to any one of claims 1 to 5. The invention according to any one of claims 1 to 6, wherein an auxiliary function is performed when the activity of the lateral effective signal source (24) around the hearing aid system (2) is determined. the method of. The method according to claim 7, wherein an appropriate hearing aid program is selected by the auxiliary function depending on the actual hearing aid situation. The signal processing apparatus (14) generates an output signal for signal processing for at least one parameter, depending on at least one parameter value, and the auxiliary function causes at least one of these for the actual hearing aid situation. The method of claim 7, wherein the parameter values are matched. The method of claim 9, wherein beamforming is performed based on at least one of these parameter values. The method of claim 7, wherein the auxiliary function determines the position of the active signal source (24) on the side relative to the hearing aid system. A first input converter (10), a second input converter (16), and a signal processing device (14) are provided, and the signal processing device (14) has at least one operation mode according to any one of claims 1 to 11. A hearing aid system (2) provided for carrying out the method described in paragraph (1). The first hearing aid (4) is provided, and the first input converter (10), the second input converter (16), and the signal processing device (14) are elements of the first hearing aid (4). The hearing aid system (2) according to claim 12. The hearing aid system includes a first hearing aid (4) and a second hearing aid (6), and the first input converter (10) and the signal processing device (14) are elements of the first hearing aid (4). The second input converter (16) is an element of the second hearing aid (6), and the element of the second hearing aid (6) is for communicating with the first hearing aid (4) and the second input signal ( The hearing aid system (2) according to claim 12, wherein the ES2) is provided for transmitting the ES2) to the first hearing aid (4).

Images